Cervical carcinoma: efficacy of thin-section oblique axial T2-weighted images for evaluating parametrial invasion.
BACKGROUND: To investigate the efficacy of thin-section oblique axial T2-weighted images in the assessment of parametrial invasion by cervical carcinoma. METHODS: One hundred parametria of 50 patients with cervical carcinoma were evaluated with pathologic correlation. We compared the sensitivity, specificity, and diagnostic accuracy in the assessment of parametrial invasion by cervical carcinoma between axial T2-weighted images and thin-section oblique axial T2-weighted images. RESULTS: Thin-section oblique axial T2-weighted images provided accurate cross sections of the cervix with excellent detail and detected parametrial invasion more accurately than did axial T2-weighted images showing cross sections of the trunk. Although the sensitivity, specificity, and accuracy for parametrial invasion were 46.4%, 91.7%, and 79.0%, respectively, on axial T2-weighted images, the corresponding values were 67.9%, 97. 2%, and 89.0%, respectively, on thin-section oblique axial T2-weighted images. There were statistically significant differences in the sensitivity (p = 0.014), specificity (p = 0.046), and accuracy (p = 0.002) in detecting parametrial invasion between these two types of images. CONCLUSIONS: Thin-section oblique axial T2-weighted images are useful for the assessment of parametrial invasion by cervical carcinoma.